Rethinking market regulation: a proposal to integrate the national
accounts and expose foreign exchange risks
This paper contends that the System of National Accounts (SNA) and Balance of Payments (BOP) are
not ‘fit for purpose’. Their usefulness is eroded because there are major differences in reporting
between stocks and flows: stocks are reported net, flows are reported gross; stocks are valued at
end-period exchange rates, flows are valued at transaction date exchange rates; there is inadequate
reporting of domestic ‘buffer’ stocks such as property, land, labour and environmental resources;
the accounts are obfuscated by financial innovation; legal entities are hidden in complex ownership
structures; and domestic sectoral balances are not confirmed with the foreign sector.
Instead, the SNA, BOP and International Investment Position (IIP) rely heavily on market pricing and
‘balancing transactions’ between the current, capital and financial account. The market risk from
these ‘balancing transactions’ would be better revealed if the stocks were reported gross and
reconciled with the flows: a form of ‘trade confirmation’ that would reveal gaps and discrepancies
between accounts from different countries.
A ‘trade confirmation’ would report stock changes on the transaction date. The design could be
adapted from database technologies where a ‘two-phase commit’ gives an indication of the quality
of the data. The publication of ‘data mappings’ would allow researchers to mix-and-match data from
different sources. These proposals become more feasible after the introduction of the Product
Identifier (PI) and Legal Entity Identifier (LEI).
Better integration of the SNA and BOP would allow researchers to study new questions and identify
foreign exchange risk in specific sectors. ‘Data mappings’ would make it easier for economists to
validate previous research findings, bring them up-to-date with new data and test new theories. In
turn, better data and more robust research would improve the measurement and prediction of key
indicators such as Gross Domestic Product (GDP) and the sustainability of the external position.
The difficulties presented with co-ordinating a ‘two-phase commit’ are illustrated with a brief
literature review and some contemporary examples: TARGET2 balances within the European
payments system; the challenges posed by external positions; trade imbalances; and gross domestic
product measurement errors. There is a possibility that greater transparency will accelerate the re-
balancing of trade and investment flows. Given the slow nature of large data projects, the risks from
market manipulation, unsustainable external positions, rogue financial institutions and hidden flows
to evade taxes and financial regulation are likely to remain. The alternative is stark: we can continue
to put our faith in private capital, financial and currency markets.
JEL: C82, E01
Keywords Methodology for Collecting, Estimating and Organizing Macroeconomic Data;
Measurement and Data on National Income and Product Accounts and Wealth; Crypto-currencies
Correspondence School of Management, University of Leicester, Leicester, LE1 7RH, United
Kingdom. Email: [email protected]
The author wishes to thank the Alumni of Leicester Unviersity, who provided generous financial
support to him and his family through the Alumni Schoalrship scheme
Reconciling the SNA and BOP
The suggestion for comprehensive national accounts had been put after the Great Depression:
‘My suggestion is that monetary theory needs to be based upon a similar analysis [to Keynes’ value
theory] but this time not of an income account, but of a capital account, a balance sheet. We have to
concentrate on the forces which make assets and liabilities what they are’
(Hicks, 1935, p. 12)
The quadruple-entry system for the SNA that we have today was developed by a sub-committee of
statistical experts working for the League of Nations, led by Stone and called the ‘social accounting
approach’ (Sub-committee on National Income Statistics, 1947). The economy is divided into eleven
sectors plus the rest of the world: households, non-financial corporations, non-financial non-farm
private enterprise, farms, the federal government, state and local government, security and realty
firms, life insurance companies, other insurance carriers, the banking system and miscellaneous
financial enterprises (Copeland, 1949, p. 257). This led to the United Nations (UN) proposing ‘A
System of National Accounts’ (United Nations, 1953). The SNA shows the balance sheet on a net
basis at the end of the accounting period: such as currency, deposits, accounts receivable and
payable, loans, equities and securities. Each stock represents both an asset and liability. Currency is a
liability of the central bank and an asset of other sectors; deposits are a liability of private banks and
an asset of other sectors; loans are a liability of the borrower and an asset of the lender; and so on.
The quadruple-entry system for the SNA arises because ‘most transactions involve two institutional
units. Each transaction of this type must be recorded twice by each of the two transactors involved…
the principle of quadruple entry accounting applies even when the detailed from-whom-to-whom
relations between sectors are not shown in the accounts’ (Various, 2008, p. 21).
In parallel, the International Monetary Fund (IMF) devloped the first ‘Balance of Payments Manual’
in January 1948 (International Monetary Fund, 2009, p. 3) The Balance of Payments (BOP) records
trade flows between residents and non-residents under a ‘double-entry accounting system… each
transaction is recorded as consisting of two entries and the sum of the credit entries and the sum of
the debit entries is the same’ (International Monetary Fund, 2009, p. 9), unlike the quadruple-entry
SNA system. Significantly, ‘while stocks are valued at end-period exchange rates, flows are recorded
on the basis of transaction rates or rates which are closer to transaction rates’ (Reserve Bank of
India, 2010, p. 14).
The BOP is a report of gross trade flows on the transaction date, where a ‘transaction is an
interaction between two institutional units that occurs by mutual agreement or through the
operation of the law and involves an exchange of value or a transfer’ (International Monetary Fund,
2009, p. 29). Each flow is income for one sector and expenditure for another: wages, interest
payments, dividends, rent, insurance premiums and benefits, public spending, taxes, subsidies and
profits. Since the SNA is reported on a net basis, investment income ( ) is also
recorded on a net basis in the BOP. These timing and reporting differences between stocks and flows
give rise to reconciliation discrepancies, not only between countries but within the datasets for the
same country. Theses discrepancies are resolved by including ‘balancing transactions’.
The sixth edition of the ‘Balance of Payments and International Investment Position Manual’ was
designed to learn ‘from the financial crises of 1994, 1997 and 1998… external debt, reserves and
financial derivatives and other leveraged and complex transactions will be scrutinised in greater
detail’ (Wang, 2005, p. 60). The intention was to fully harmonise the SNA and BOP ‘on issues such as
the defining resident units (producers or consumers), valuation of transactions, the stock of external
assets and liabilities, time of recording transactions and stocks, conversion procedures, coverage of
international transactions in goods, services, income, capital transfers, and foreign financial assets
and liabilities’ (Reserve Bank of India, 2010, p. 19). This harmonisation resulted in the inclusion in
BOP fifth edition of the International Investment Position (IIP), a stock position, although this is only
a ‘subset of the assets and liabilities in the national balance sheet’ (International Monetary Fund,
2009).
In this theoretical framework there are four sections: the current, capital and financial accounts and
the official reserve assets, which sum to zero according to this accounting identity:
Equation 4.1:
Where CAB is the current account balance (goods and services, plus net income); CAP is the balance
on the capital account; FAC is the balance on the financial account; and ORT are the official reserve
assets.
The current account comprises imports, exports, net income on domestic investments and net
income on foreign investments:
Equation 4.2: `
Where I = imports; X = exports; B£.r£ are the net income on domestic investments; and B$.r$ are the
net income on foreign investments.
The capital and financial accounts comprise changes in holdings of foreign and domestic
investments, and other capital transfers:
Equation 4.3:
Where ∆B£ are net changes in holdings of domestic investments; ∆B$ are net changes in holdings of
foreign investments; and ∆CAP are other capital transfers
Lastly, the official reserve assets are mainly comprised of gold and foreign currency assets:
Equation 4.4:
Where ∆G are changes in gold and ∆O$ are changes in foreign currency assets.
Tables 4.1 and 4.2 show a high-level ‘data mapping’ between the SNA and BOP. Note that different
names are given to each section; the domestic sector is further disaggregated in the SNA; timing
differences give rise to valuation differences; parts of the balance sheet are missing, incomplete or
optional; and the SNA is reported net whereas the BOP is reported gross:
Table 4.1: Reconciling the SNA and BOP
System of National Accounts (SNA) Balance of Payments (BOP)
GDP(E) Gross domestic product 1700
=G Government final consumption 400
+C Non-government final
consumption
800
+I Gross capital formation 400 Goods and services
+X Exports of goods and services 500 Credit (export) 500
+M Imports of goods and services -400 Debit (import) -400
X-M Trade balance 100
Current and capital account
Current income
Credit 100
Debit -50
+B£.r£
+B$.r$
Net current income from abroad 50 Total 50
Current transfers
Credit 300
Debit -150
Net current transfers from abroad 150 Total 150
CAB Balance on current account 300
GNI Gross National Income 1900
GNI = GDP +B$.r$ +B£.r£
Gross saving ( = GNI – C + G ) 700
Adapted from (Reserve Bank of India, 2010, p. 21)
Table 4.2: Gross Saving, Capital and Financial Accounts
System of National Accounts (SNA) Balance of Payments (BOP)
Gross saving ( = GNI – C + G ) 700
Capital account
+∆CAP Net capital transfers from abroad 5 Credit 5
Net acquisition of non-produced, non-
financial assets
-15 Debit -15
Less gross capital formation ( - I) -400 Total -10
Net lending (+)/net borrowing (-) 290
Financial Account
Assets 400
Liabilities 110
+∆CAP Net lending (+)/net borrowing (-) 290 Total 290
Adapted from (Reserve Bank of India, 2010, p. 21)
This brief overview illustrates that account reconciliation across time, countries and sectors is a
difficult process. The evolution of different versions means that older datasets and research cannot
be readily compared.
The implication is that ‘balancing transactions’ are settled via market processes: in capital, financial
and currency markets. Real goods and services are exchanged for financial and capital assets,
including new classes such as derivatives. If market processes fail, the government or central banks
must step in to restore order: expanding or contracting domestic and foreign reserves, devaluing the
currency, or writing off debts.
In the SNA and BOP, ‘balancing transactions’ are reported ex-post as a residual, after accounting for
‘any valuation changes such as that caused by changes in exchange rates and other adjustments’
(Wang, 2005, p. 60-61). In other words, the winners and losers in foreign exchange and financial
markets are hidden from view. Secondary markets, and leveraged investment in particular, obscure
the underlying accounts further.
To illustrate this, consider the stylised flows in Figure 4.1. where Country A earns all of its foregn
income from the export of goods and services; Country B runs a trade deficit by exporting financial
assets; and Country C earns investment income by speculating on foreign assets. Lastly, Country C is
the ‘reporting currency’ for both the SNA and BOP, so stocks and flows are recorded in currency C:
Figure 4.1
The BOP shows gross trade on the relevant transaction dates, shown in BLUE and recorded in foreign
currency C. The BOP also shows, in RED, net investment flows. However, there is no record of the
currency transactions, shown in YELLOW.
With currency transactions missing, the BOP cannot be reconciled with the SNA. In this example,
Country C was leveraged in the middle of the period, but at year end the financial account only
shows a net transfer of financial assets from Country B to Country A: leverage obscures the
intermediate financial flows.
Figure 4.2 shows the same transaction, but recorded as integrated National Accounts. In this
example, financial and capital flows are recorded on the transaction date, in the domestic currency.
The gains and losses in currency and financial markets are exposed:
Figure 4.2
Product and legal entity ledgers
In other spheres of life, such as retail, technology has been employed to trace every movement of
physical goods from the manufacturer, through suppliers and retailers, to the consumer. Recent
proposals to introduce a legal entity identifier (LEI) and product identifier (PI) to trace financial
products (Ali, Haldane, & Nahai-Williamson, 2012) have seen the G20 agree to implement a global
LEI that ‘uniquely identifies parties to financial transactions’ (Financial Stability Board, 2012, p. 24).
In terms of consumer goods, the PI is the equivalent of the barcode and the LEI identifies the buyer
and seller.
Like consumer goods, the re-combination of components makes financial assets difficult to
categorise. The key attributes go beyond the initial amount, market value and payment streams.
Depending on the buyer and seller, there are multiple risk and return streams: not just currency and
country risk, but default risk, counterparty risk, duration risk and so on. Some asset categories bring
additional complications, with fire, theft, disaster, health and environmental risks. In turn, these
additional risks can be aggregated, disaggregated, hedged, insured, or used as a basis for
speculation.
There are similar complications with the LEI. The control of transnational corporations is highly
concentrated, with ‘nearly 4/10 of the control… held, via a complicated web of ownership relations,
by a group of 147 TNCs in the core, which has almost full control over itself’ (Vitali et al., 2011, p. 4).
The recursive and concentrated nature of this hierarchy suggests financial fragility, but the data are
difficult to analyse. Lastly, ownership of financial assets is often fractional and highly liquid, with
rapid turnover in secondary markets due to financial innovation: exchange-traded (ETF) which create
pooled funds where individuals trade fractional shares; high-frequency trades where ownership
changes in less than a second; derivative trades where synthetic assets and liabilities can be
swapped, mature and settle before the balance sheet is published.
The ability to have international ‘trade confirmation’ depends heavily, therefore, on the pre-
existence of an international product and legal entity register: who (the sectors) are the parties to
the trade, what (the product) is being traded, and when. An analogy with internet protocols might
be useful. The nodes (sectors) are defined by a communication protocol called TCPIP (transmission
control protocol). This standard affords some anonymity to the user, who is known by an IP address.
Internet protocols also define the message (the products) and give an international timestamp to
each message (when).
Anonymity can be extended because technology companies build their own protocols on top of the
core protocols. In other words, it is possible to design a local system that interfaces with the
international network using a ‘data mapping’, that further enhances your local anonymity. By
offering translation services from the local system to the international system, that is compatible
with the TCPIP protocol, it is still possible to logon to the internet with the simplest of computers,
running earlier software versions, and link to modern protocols. In other words, protocols provides
interoperability across space and time, anonymity, and localization.
If the LEI were to match this level of sophistication, economists would be able to integrated National
Accounts in the same way. The provider (such as the Office for National Statistics) would present
whichever region and year(s) the economist asked for, using whichever ‘data mapping’ the
economist required and was supported by the core protocols. It would be possible to re-create the
empirical work for an old research paper; to re-run empirical work with more recent data; and to see
the impact of changes to the protocols on that empirical work.
Defining the core protocol is far from simple, but design patterns from database systems might be
useful. At present, the SNA and BOP are reported ex-post and it is quite possible for two countries to
report completely differently on the same transaction: in database terms, this is a ‘conflict’ and the
most common solution is to report a ‘balancing transaction’, such as the net errors and omissions
outlined earlier.
The SNA and BOP could learn from database systems. In a ‘two-phase commit’ each participant
reports to a co-ordinator on whether they agree or disagree with the transaction: the co-ordinator
decides whether to confirm or roll-back. There are other transaction patterns, such as the ‘three-
phase commit’, where the transaction proceeds because it is confirmed or because the other party
has timed out. The advantage of confirming the trade in this way is that each data item has a
measure of certainty or quality associated with it. So the US trade deficit might be confirmed by all
trading partners, or with a proportion, or with none.
So this prposal is for two simple recommendations based on the core PI and LEI protocols. The first is
for the data providers (IMF, OECD, UN, ONS and so on) to provide ‘data mappings’ that translate the
SNA and BOP across countries and time: the researcher knows if they have chosen the SNA 1993 or
2008 presentation of the data. This allows researchers to recreate historic research; bring old
research findings up to date; and reveal the impact of new data standards on historic research. The
second is for the data aggregators or co-ordinators to move towards ‘trade confirmation’. As a
minimum, each country would need to use the same public ledger to record contemporaneous
transactions. They might be aggregated and anonymised locally, but the public ledger would
represent the minimum information needed for economic research: asset and liability types (what),
country, sectors and sub-sectors (who). The existing SNA and BOP include much more detailed
sector and product decomposition.
TARGET: a model, or a flawed implementation?
The Trans-European Automated Real-time Gross Settlement Express Transfer system, or TARGET,
includes some of these features. TARGET is the payments system for European central banks, and is
a platform for multiple central banks and private banks to report cross-border settlements as
TARGET imbalances. The nature of these imbalances is the subject of detailed economic research,
such as (Cecchetti, Mccauley, & Mcguire, 2012; H.-W. Sinn & Wollmershäuser, 2011). Unlike the
more general design in Figure 4.2., however, TARGET does not incorporate foreign exchange
flexibility: the unit of account (the Euro) is fixed.
The TARGET system is not ‘subject to any legal netting or set-off arrangement’ (Bank of England,
2000, p. 74). In other words, the reserve balances in Equation 4.1 continue to grow in parallel with
net imbalances across the capital and financial accounts. TARGET acts much like a large, pooled fund
of reserve assets and liabilities, comprising of Euro-denominated securities and deposits which are
‘predominantly matched by euro bills together with currency, foreign exchange and interest rate
swaps, which are off-balance sheet’ (Bank of England, 2000, p. 71).
As a settlement system, TARGET does offer a basic mechanism to report currency flows between
countries, the YELLOW transactions in Figure 4.2. The difference is that the Euro project went much
further by denominating all currency flows in Euros: there is no ‘balancing transaction’ in foreign
exchange as a consequence of the Deutschmark being strong, the Drachma being weak, and so on.
Researching central bank balance sheets under floating currency arrangements, to fully understand
the winners and losers in foreign exchange markets, is rather hit-and-miss. The central banks of
Russia and Europe have produced English-language Annual Reports since 1998 and 1991
respectively. Since March 2013, the People’s Bank of China has provided monthly data (People’s
Bank of China, 2012), but like most central banks does not publish the currency composition of
foreign reserves. The Swiss National Bank has published the currency composition since 2008, but
the size of their foreign reserves is tiny relative to the others (Swiss National Bank, 2013a).
A possibility for this kind of research is to use the International Financial Statistics (IFS) or the
Currency Composition of Official Reserve Assets database (COFER), which countries subscribe to
voluntarily and are collated by the IMF. The historic data are incomplete: the Swiss National Bank
are available from 1993; the Bank of China in 1994; the Bank of England in 1996; the Euro area in
1997; and the Federal Reserve, Bank of Japan and Russian Federation in 2001. At the aggregate level
COFER reports on the currency composition of reserve assets, but only at the aggregate levels of
‘advanced’ and ‘emerging and developing’ and only for five major currencies: the US Dollar, Euro,
Sterling, Japanese Yen and Swiss Franc. COFER currency compositions for individual countries are
not published and the reserves are converted to US Dollars and only exist back to 1995 (annually)
and 1999 (quarterly).
So, again, there are problems with using this data for research: it is not disaggregated by currency,
and it is difficult to map from one period to another such as the introduction of the Euro. To
illustrate the problem with ‘data mappings’, TARGET balances are categorised as ‘foreign assets’
which are ‘those external assets that are readily available to and controlled by monetary authorities
for meeting balance of payments financing needs, for intervention in exchange markets… and for
other related purposes (such as maintaining confidence… (and) as a basis for foreign borrowing)’
(International Monetary Fund, 2009, p. 111). If they are held by a quasi-public body, like a Sovereign
Wealth Fund, they might not be classified as ‘foreign assets’ because sovereign wealth funds and
national investment banks are not considered to be sufficiently ‘liquid’ for monetary purposes.
There is also a ‘residency concept’, defining ‘foreign assets’ as a claim on non-residents.
Table 4.3 shows the IFS data categories. On the liability side, there are reserve deposits and
currency. On the asset side, there are foreign reserve assets, and claims on banks, non-banks, the
private sector and governments:
Table 4.3
International Financial Statistics
Assets Foreign assets These are the net claims on non-residents in
monetary gold, SDR holdings, foreign
currency, deposits , securities (other than
shares), loans, financial derivatives and other
Claims on domestic
banking institutions
These are the net claims on depository
corporations
Claims on non-bank
financial institutions
As above, but for non-banks
Claims on private sector As above, but for the private sector
Claims on domestic
government
These are generally in the form on
government securities
Liabilities Reserve deposits In return for selling an asset to the central
bank, reserve deposits are created. Banks
must hold a reserve buffer, but excess reserve
deposits are available to buy assets
Currency in circulation Notes and coins
Sources: (International Monetary Fund, 1993; International Monetary Fund, 2008)
Table 4.4 then maps the IFS data categories with Bank of England data categories for 2000:
Table 4.4
Bank of England Balance Sheet (as of 29th February 2000) Total
IFS Assets Foreign assets £19.9 billion
Claims on domestic
government
£3.9 billion £23.8 billion
Liabilities Reserve deposits £25.1 billion £25.1 billion
Bank of
England
Accounts
Assets TARGET £13.7 billion
Debt securities £3.5 billion
Loans and advances £5.4 billion
Other £1.3 billion £23.9 billion
Liabilities TARGET £13.6 billion
Deposits by central banks £3.0 billion
Eurobills £2.1 billion
Other £5.2 billion £23.9 billion
Source: own data
This example illustrates the two problems outlined. There is no publication of ‘data mappings’
between providers and, although the publication by the Bank of England of the TARGET balance as
an asset type is useful, it is not decomposed by country or exchange rate (we can at least assume the
balances are within the banking sector). Consequently there is no breakdown in the Bank of England
report to show the TARGET profits and losses disaggregated by country: TARGET acts like a single,
pooled fund.
Making reserve decompositions a public ledger might encourage further netting between
counterparties, or it might encourage speculation against the central bank, or even entire regional
banks. Netting might be more likely if there were dis-incentives for large credit and debit balances,
but no such Pigovian taxes exist. On the other hand, publishing the public ledger would allow
researchers to study, in more detail, the causal relationships between bank balance sheets, trade in
goods, and trade in financial assets. Given the lack of macroprudential dis-incentives to large
imbalances, it might be prudent to release historical datasets first.
Despite the rigidities from having a fixed exchange rate with other countries, TARGET illustrates the
technical advantages of a ‘two-way commit’. There is no room for accounting holes in the banks
balance sheets: every seller must have a buyer and vice-versa. There is therefore greater
transparency to off-balance sheet vehicles. This need for transparency applies to central banks as
well as private banks: for example, the Bank of England Asset Purchase Facility Fund Ltd (BEAPFF)
was treated as ‘off-balance sheet’ in the Bank of England’s annual report (Bank of England, 2012d, p.
69), and only the loan from the Bank of England to the BEAPFF was reported. The assets would be
missing from the UK balance sheet if BEAPFF did not report as if it were a private UK entity.
As it happens, the BEAPFF appears on the government balance sheet in the ‘Blue Book’ as an asset
adjustment: in 2009, UK Gilts increased by £209 billion and short-term loans by UK MFIs decrease by
a similar amount (Office for National Statistics, 2012b, p. 176 - 179). So, even between national
reporting bodies, such as the ONS and Bank of England, there are discrepancies.
Financial and capital account imbalances
There are already some datasets that show the research value of a public register of private assets.
The US Treasury publishes foreign holdings of US securities, and these hint at the major imbalances
in international trade in goods as well as areas of wealth accumulation. In September 2013, most
foreign holding of US securities were in Japan then China (both at $1.8 trillion); then the United
Kingdom, Cayman Islands and Caribbean (US Department of the Treasury, 2013). These country
positions are qualitatively different. In Japan, it is mainly private citizens who hold US treasuries; in
China it is mainly the central bank which also holds US treasuries; and in the three financial centres
the majority are private holdings of corporate securities.
Extending this public register to other countries and asset classes is essential data for research into
the sustainability of the external position of countries like the US and UK. Some academics, like
Ricardo Hausmann and Ferdico Sturzenegger, have argued that the external position has nothing to
do with leveraged investment (Figure 4.2) but, rather, the higher yield on foreign assets, compared
to foreign loans, is a consequence of ‘dark matter’ and the mis-measurement of the balance sheet.
'Dark matter' is defined according to the yield differential, so assumes that the investment income is
a more accurate measure way to estimate stocks than the direct measurement of the balance sheet,
which does not include 'surprises, risk premia and embedded services [insurance and liquidity]'
(Hausmann & Sturzenegger, 2006, p. 6). Critics say that ‘the dark matter view fails, as it rests on an
assumption that income streams are the most accurate items in the entire set of international
accounts. Given that the bulk of income streams are not measured but are formed by applying
estimates to estimates, this assumption is false’ (S. E. Curcuru et al., 2008, p. 19).
An alternative view is that countries like the UK and US have an ‘exorbitant privilege’ that is mostly a
consequence of trade in short term liquid assets, a conclusion that has been reached by several
other authors (Gourinchas & Rey, 2007; Lawrence, 1990). This alternative perspective includes the
idea of a tipping point, beyond which the economy is unstable and the net yield on the external
liabilities exceeds the net yield on external assets:
Equation 4.5:
Where A = net foreign assets, L = net foreign loans, = return on assets; and = interest rate on
loans
There may also be different tipping points within each sector and asset class, where the returns on
foreign assets must be higher than the cost of financing. This is one area of research which would
benefit greatly from disaggreggating the balance sheet by sector, to show who is borrowed from
whom:
Equation 4.6: ∑ ∑
Adapted from (Gourinchas & Rey, 2007, p. 27)
Gourinchas and Rey have suggested the tipping point for the US economy will be when liabilities are
1.43 times assets, a figure estimated using past returns and assumptions such as ‘most US liabilities
are in dollars, whereas a share of US assets are in foreign currency’ (Gourinchas & Rey, 2007, p. 29).
However, this analysis relies on extrapolation of the yield differential. A domestic central bank has
no policy lever to push up the yield on foreign assets. Similarly, with a persistent trade deficit,
domestic assets are being exchanged for (zero yield) goods, not high-yielding assets. If foreign goods
are swapped for domestic assets, the swap is (positive yield) assets for (zero yield) goods, and
further risk of reaching a tipping point.
The tipping point might be avoided by running a current account surplus: in goods, services and net
investment flows. Active speculation on the financial account, buying high-yield assets and selling
low-yield assets, is another strategy, as is selling the ‘lemons’ (Akerlof, 1970) to non-residents.
However, the currency composition of the external position matters a great deal: having a flexible
exchange rate is a safety valve only if foreign liabilities devalue more quickly than foreign assets. If
not, re-balancing is only possible via the trade channel. Currency devaluation has other risks,
including exchange rate passthrough as inflation pressures.
The risk is that being a currency issuer and speculating on the financial account are insufficient to
prevent reaching a tipping point. Being a sovereign currency issuer gives public assets ‘other virtues
(eg: safety) which makes them attractive in spite of their lower yield' (Hausmann & Sturzenegger,
2006, p. 4). Still, Equation 4.7 underlines the exacerbating effect of a trade deficit. If capital and
financial assets are being sold, the economy is exchanging a higher yielding asset for goods:
Equation 4.7:
If you are unable to pay for foreign goods with domestic currency, this exacerbates capital flight.
Equation 4.1 suggested two solutions. One is to accumulate foreign currency reserves during good
times, that you can sell during periods of capital flight. The second is to sell capital (non-produced,
non-financial) assets: the Washington Consensus solutions to privatise public assets, and make
existing private assets as attractive as possible to non residents by driving domestic wages down,
and foreign yields up.
Understanding these tipping points is also obfuscated by incomplete coverage of the balance sheet:
not just of derivatives, but property, labour and other categories, despite the existence of ‘well-
organized markets... to facilitate the buying and selling of many kinds of existing fixed assets, notably
automobiles, ships, aircraft, dwellings and other structures’ (Various, 2008, p. 198). Immovable
assets owned by non-residents are ‘deemed, by convention, to be owned by resident units’ (Various,
2008, p. 201) yet they result in an international currency flow. A persistent trade deficit might be
funded by selling a steady stream of these hidden assets, or by intra-firm transfers of such capital as
intellectual property, software, machinery, goodwill, art and entertainment rights.
These examples show that the composition of the balance sheet can have quite unpredictable
effects on the sustainability of the external position. There are one-off boosts from the sale of
property, privatisations, mineral rights and land, but there are long-term consequences in terms of
the flow of rents, interest payments, profits, wages, royalties and licence fees. These are all areas
where integrated National Accounts will improve understanding.
International trade imbalances
The second area where better data are needed in to understand the dynamics between
international trade and reserve accumulation. From 2000 to 2011, the official foreign assets of
central banks rose from six per cent to over 14 per cent of World GDP, with China accounting for
over 50% of the total for ‘emerging and developing’ countries. As Figure 4.3 shows, China’s official
reserves are much higher than the current account surplus:
Figure 4.3
Sources: COFER, Bank of China; State Administration of FX, China; WM Reuters
Critics argue that that China’s trade flows are incorrectly reported, with argument for over-invoicing
to borrow cheaply in US Dollars and invest in renminbi (Altman, 2013), and for under-invoncing to
avoid trade tariffs (Ferrantino, Liu, & Wang, 2012). Other explanations for high levels of reserves
look to the financial and capital account, such as high levels of non-US foreign direct investment in
China (Xu, Lin, & Sun, 2010). This research would benefit greatly from a ‘two-way commit’ between
exports and imports: checking the goods out and checking them in at the other border.
There is evidence that China’s capital and financial account are relatively open, in that currency
traders describe the offshore renminbi market as being fairly liquid:
“The Chinese currency, or you know, Korea, India, Taiwan, you know, Brazil, these sorts of
countries... they’re liquid but they’re not accessible… we use, we do access... you can’t actually
access the on-shore market because of capital controls… I don’t know if you’re aware of non-
deliverable forwards (NDF) … it’s essentially a US Dollar settled contract for difference that’s traded
over-the-counter with another party… it just depends I think how porous the capital controls are… In
China it’s less reliable, but still pretty reliable… but countries like, take, Brazil and Russia the NDF is
as good as being exposed to the onshore”
2012 interview with UK FX manager SPACEMAN
Indeed, China has seen FX turnover in renminbi grow from about $1 billion in April 2004 to $20
billion in April 2010 (BIS, 2010, p. 82) and to $120 billion by April 2013 (Bank for International
Settlements, 2013, p. 13). Estimate of offshore renminbi deposits are around 120 billion US dollars in
Hong Kong and London (Bourse Consult, 2012, p. 17), and are increasing rapidly.
Apart from trade in goods and services, foreigners have been able to buy ‘B’ and ‘H’ equity shares in
Chinese companies since 1992, and Qualified Foreign Institutional Investors can invest directly in ‘A’
equity shares: these holdings are estimated to be around 5-6% of the total (Chen, Du, Li, & Ouyang,
2013, p. 661). Since the capitalisation of the Shanghai and Shenzen stock exchanges is around 4
trillion US dollars: that accounts for approximately 200 billion US Dollars.
A third driver of reserve accumulation has been net investment income on the bonds already
accumulated. The total return on a US benchmark bond was about 31% from 2001 to 2007, where
total return is the combination of capital gain and the coupons.
A fourth possibility is that the People’s Bank of China made significant profits on gold reserves, with
a much higher rate of return. Gold bullion, for example, returned almost 600% from 2001 to 2013
when it reached 1,600 US Dollars per ounce. Although data are scarce, official godl reserves in China
are less than 2% of the total (Bloomberg News, 2013; People’s Bank of China, 2012).
Together, the current account surplus, gold profits, sale of financial assets and offshore renminbi do
not fully account for the reported reserves: there is a discrepancy of at least $300 billion:
Table 4.6
Item (in billions USD) 2001 2007 Gain
Reserves 240 1,599 1359
Less
Current account surplus (billions USD) -37 -353 -316
Offshore renminbi (billions USD) -120 -120
Equity investments (billions USD) -200 -200
Subtotal 203 926
Discrepancy 723
Less 31% capital appreciation on T bills -369
Less tripling of 2% gold reserves -60
Discrepancy -$300
Sources: Datastream, own calculations
Large accumulations of foreign reserve assets have also been the case at the Swiss Central bank, in
the Euro area, and in the other BRIC economies (Brazil, Russia and India). Another possibility is
simply that private investors began to shun financial assets from certain developed countries, selling
them on international exchanges and swapping foreign currency for local currency. At the aggregate
level, the central bank, this shows up as an accumulation of foreign reserves if the developed
country is also running a trade deficit:
Table 4.7
Foreign Reserves (in Billions)
Annualised
Growth (2001-
2012)
December 2001 December 2012
In Local
Currency
In USD (at
end of year
spot rate)
In Local
Currency
In USD (at
end of year
spot rate)
Swiss National
Bank
88 53 491 536 23%
Federal Reserve 68 68 160 160 8%
Euro Area 399 355 800 1,055 10%
Bank of England 6.7 9.8 9.13 14.8 4%
Bank of Japan 4,250 32 5,750 69.7 7%
Brazil 63,248 30 756,926 416 27%
Russia 888 31 16,187 551 30%
India 2,326 50 15,697 309 18%
China 1,986 240 24,142 3,828 29%
Sources: International Financial Statistics, WM Reuters
In other words, there is a much simpler explanation, with large shifts in international investment
portfolios that were not immediately apparent because there were no price effects. For that to be
the case, the sell off in some countries would need to be matched by leveraged buying in other
countries. Understanding these nature and composition of the balance sheet would, perhaps, help
to guide some of the high profile politics. The Treasury Secretary, Timothy Geithner, has interpreted
reserve accumulation as evidence that ‘China manipulates its currency… (which) should be resisted
through protectionist policies’ (Dooley, Folkerts-landau, & Garber, 2009, p. 5). In October 2011 the
US Senate passed the Currency Reform for Fair Trade Act to permit countervailing duties (tariffs)
against any foreign country with a ‘fundamentally undervalued currency’(Levin, 2011). At the same
time, US and UK banks and private interests have been accused of manipulating capital and currency
markets, including interbank rate manipulation (Kregel, 2012) and exchange rate manipulation
(James O’Toole, 2013).
To balance this perspective, the Independent Evaluation Office (IEO) of the IMF has stated that
official reserve assets ‘remain small relative to global banking assets which themselves have
experienced a leverage induced "global banking glut" (Independent Evaluation Office, 2012, p. 8). By
2010, assets held by domestic and offshore financial centres had grown to over 25 times the size of
official reserves (Independent Evaluation Office, 2012, p. 8), or about four times the world’s gross
domestic product. That compares to the official reserve assets which are about one tenth of the
world’s gross domestic product: a ratio of 250:1. This example suggests that a ‘two-way commit’
would not only improve economic research into stability concerns, but it would help exporters and
importers collect taxes and tariffs, and it would raise the quality of the policy debate.
GDP measurement errors
As well as the issues of current and financial account sustainability, data problems give rise to
estimation errors of indicators such as gross domestic product. There are are, in fact, three
measures of gross domestic product. Table 4.1 showed the expenditure approach, GDP(E), which is
the sum of all final expenditures within an economy, plus gross capital formation, plus the trade
balance:
Equation 4.8: GDP(E) = G + C + I + ( X – M )
There are also an income approach, GDP(I), and a production approach, GDP(P). GDP(I) is the sum of
all incomes directly generated by productive activity. In other words, the income of government
employees is excluded. GDP(P) is the sum of all (estimated) production activity from both goods and
services:
Equation 4.9: GDP(I) = compensation + gross operating surplus + mixed income + net taxes
Equation 4.10: GDP(P) = output – intermediate consumption + net taxes
Adapted from (Office for National Statistics, 2012b, p. 2)
Where compensation includes pensions; operating surplus includes rental income; mixed income
includes the income of the self-employed; and net taxes are taxes on production and products less
any subsidies
All three approaches to GDP are, therefore, subject to errors in the measurement of ‘balancing
transactions’. If final consumption is estimated from consumption taxes, and exports and imports
from trade statistics, then the investment figure in GDP(E) is a residual and subject to the
measurement error in ‘balancing transactions’. Similarly, GDP(I) can be inflated or deflated by
inaccurate measurement of the gross operating surplus, and GDP(P) relies on accurate measurement
of output. Table 4.1 shows that GDP measurements would be more accurate if financial and capital
account transactions were known with more certainty.
The impact of the capital and financial account on gross capital formation can be quite variable.
Consider some contemporary examples in the UK: the privatisation of the Post Office, and the sale of
a central London property to a foreign investor. Under privatisation, services provided by the Post
Office are recategorised without affecting the overall number. However, privatisation affects the
balane sheet: private borrowing rises and government borrowing falls. The Post Office can sell
capital assets to a foreign entity and lease them back, which reduces private borrowing but has a
negative impact on the current account since investment income flows abroad.
The re-categorisation of compensation from public to private sector can have significant effects on
taxation if employment is relocated abroad. Ceteris paribus, the effects of privatisation depend on
the interplay between taxes and investment income flows. The worst case for GDP is that fewer
domestic taxes are paid, and investment income flows to foreign rather than domestic investors. On
the other hand, if the privatised firm pays more taxes, or the government sells a loss-making service,
privatisation implies a boost to GDP. Either way, the government has swapped uncertain changes to
tax and investment flows for a certain reduction in total debt servicing costs. Taken together, this
swap suggests more volatility in GDP.
For the swap to be neutral, the government would need to reduce spending if taxes from the
privatized Post Office fell, or if investment income flowed offshore. Higher volatility in GDP also
comes at a cost to government.
The sale of a UK property to a non-resident investor also impacts GDP. The first thing to note is that
‘despite their obvious importance, even stocks of residential dwellings are not publicly available for
more than a handful of countries’ (Various, 2008, p. 404). In other words, the balance sheet is
incomplete, and the initial sale only impacts GDP if there are secondary effects: higher domestic
consumption, or even an increase in investment income if the resident invest in foreign assets.
The other impact is from the actual use to which the property is put. If the property is rented out,
and the non-resident transfer the rent abroad, there is a decline in GDP; if the proceeds are spent in
the UK on property there are secondary effect on GDP; and so on. In both cases, a ‘two-phase
commit’ between residents and non-residents would shed some light on the effects of privatization
and property sales.
Building on crypto-currency design
The hidden exchange of currency and commodities, such as crypto-currencies, makes a ‘two-phase
commit’ difficult because the legal entities are anonymised. Large denomination bank-notes (and
gold) are also difficult to trace, although more difficult to move across borders than crypto-
currencies. For the US, an estimated half of the growth in currency issue since 1988 was due to
foreign holdings, with about three-quarters of currency held as the highest denomination $100 bill
(Judson & Judson, 2012, p. 8). When Euro cash was introduced in 2002, a EUR 500 denomination
note was issued; Japan has issued a YEN 10,000 note since 1958; and the highest denomination note
from among the G5 is the CHF 1,000, first issued in 1907 and amounting to 60% of the value of Swiss
notes in circulation in 2013 (Swiss National Bank, 2013b).
Yet these foreign holdings of domestic bank-notes are an exchange risk for central banks, if non-
residents decide to exchange their bank-notes en masse. According to the IFS, the Bank of Japan has
the greatest exposure in proportion to foreign assets, followed by the Federal Reserve and Bank of
England:
Figure 4.4
Sources: IFS, WM Rates, bank-notes as percentage of GDP
In fact, the public ledger adopted by crypto-currencies could be easily adapted to be both
anonymous and to have a ‘two-way commit’: the public ledger already records the ‘what’ (the
product is a crypto-currency) and the ‘when’ (a ledger). The LEI could be anonymized but still
contain met-data that identifies then accounting sector and country.
Moving towards integration
The current SNA and BOP would need to adopt a single, international, product and legal entity
identifiers (PI or LEI) as a first step towards integration. This represents a massive co-ordination
problem. At present, each country publishes their own set of national accounts, usually via the
central bank or a statistical body like the Office for National Statistics (ONS) in the UK. National
datasets are collated by supranational and international bodies, like the IMF, UN and the
Organisation for Economic Co-operation and Development (OECD): the OECD being the successor to
the organisation which ran the US-financed Marshall Plan after World War II. The flow of data is then
from national to supranational and international bodies.
The historical data series have major gaps. Flow of funds data for Europe, for example, are not
available before 2002 (Duc, 2009, p. 5) and ’reliable high frequency time series exist [only] for a
fraction of the world economies’ (Kinsella, 2011b, p. 12). Given the scale of the co-ordination
problem this is hardly surprising, despite advances in communications, data processing and data
storage and data technology. In 2009 the Bank of England reported that ‘in the UK, much work
needs to be done to fill the data gaps identified from the financial crisis. One issue is the
development of a flow of funds model for the UK’ (Murphy, Westwood, & Murphy, 2009, p. 23).
Much of the UK data are judged to be ‘poor quality’ with ‘six significant asset classes for which (there
are not) unique data on sectoral holdings, of which quoted and unquoted equity and short and long-
term debt are the most important’ (Barwell & Burrows, 2011, p. 7). Reports from the European
Central Bank, Bank of Japan, IMF and Bank of England (Duc, 2009; IMF Staff, 2009; Konno, 2010;
Murphy et al., 2010) call for improvements to the quality, timeliness and coverage of regulatory
data, in particular the need to keep up with financial innovation, notably structured credit and credit
risk products, and to better monitor the ‘interconnectedness of systemically important financial
institutions’. The top three data issues identified by the IMF were ‘aggregate leverage and maturity
mismatches… the financial linkages of systemically important global financial institutions and… cross-
border activities of nonbank financial institutions‘ (IMF Staff, 2009, p. 11). Economists who work
with the flow-of-funds data also express dissatisfaction, particularly working with countries outside
the US (Zezza, 2012).
Errors, omissions and revaluations are an intrinsic part of the reports. Table 4.8 shows that the net
errors and omissions (NEO) for the UK BOP, in 2010, were about three times the changes in the
capital account and official reserves, and around 18 per cent of the current account deficit:
Table 4.8
UK Balance of Payments (2010 Figures) £ billions
Current account (71.60)
Goods (exports) 410.22
Goods (imports) (563.15)
Balance on goods (152.93)
Services (credit) 238.75
Services (debit) (169.14)
Balance on goods and services (83.32)
Income (credit) 255.42
Income (debit) (212.94)
Balance on goods, services and income (40.84)
Current transfers (credit) 21.95
Current transfers (debit) (52.72)
Capital account 5.01
Capital account (credit) 9.41
Capital account (debit) (4.40)
Total, current and capital account (66.60)
Financial account 63.57
Direct investment (abroad) (10.67)
Direct investment (United Kingdom) 46.95
Portfolio investment assets (130.88)
Equity securities (12.59)
Debt securities (118.28)
Portfolio investment liabilities 143.94
Equity securities 3.60
Debt securities 140.35
Financial derivatives 44.94
Financial derivatives assets
Financial derivatives liabilities 44.94
Other investment assets (359.94)
Monetary authorities
General government (1.56)
Banks (212.18)
Other sectors (146.20)
Other investment liabilities 329.22
Monetary authorities
General government 1.21
Banks 96.65
Other sectors 231.35
Total, current, capital and financial account (3.03)
Net errors and omissions 13.04
Reserves and related items (10.01)
Reserve assets (10.01)
Conversion rates: GBP to USD 0.65
The final part of this paper considers the impact of a ‘two-phase commit’ and alternative mappings
on three examples: TARGET balances; cash and crypo-currencies; and the trade balance.
Recommendations
This paper proposes a public transaction ledger to reconcile international trade. The key trades that
are missing from the SNA and BOP are currency and foreign exchange: rather, the financial system
puts blind faith in trade reconciliation via ‘the market’. Yet the list of potential market failures is
growing: current account imbalances, reserve accumulation and unsustainable external positions.
These external positions include tipping points at the sectoral level (households, firms, banks and
governments) as well as at the aggregate level. Gaps in the balance sheet around derivatives,
property, land and labour exacerbate the difficulty of creating a public ledger.
The PI and the LEI enable two of the three salient features of a public ledger: ‘what’ and ‘who’.
Design patterns from database and internet technologies offer ways to deal with the key issues of
anonymity and localization. The TARGET settlement system for Eurozone countries is a very rigid
interpretation of a public ledger, with no ‘balancing transactions’ due to foreign currency
fluctuations. At the other extreme, crypto-currencies are a very loose interpretation, with a single
product (the crypto-currency), a single country and currency (the world), and a single sector.
This Chapter proposes adopting other technology patterns, in particular the publication of ‘data
mappings’ between national, supranational and international regulators. These ‘data mappings’
could show the relationships between datasets across time as well as countries, making it easier for
economists to validate previous research findings, bring them up-to-date with new data and test
new theories. With less reliance on ‘balancing transactions’ economists will have a better chance of
measuring gross domestic product and evaluating polices. Economics might become a more modest
discipline, where economists specialise by sector, country and historical period.
There are some risks from the rapid re-balancing of trade and financial flows. Gourinchas and Rey
have estimated that to return the US trade balance to equilibrium with financial account requires an
‘implausible depreciation of 75 per cent’ in one year or ’a depreciation of 18 per cent per year’ over
five years (Gourinchas & Rey, 2007, p. 32). Bulk public data collection also raises major issues about
privacy rights and government power (New York Times, 2014): these issues do not exist with
anonymous public ledgers, such as the cryptocurrency pattern. Given the slow nature of large data
projects, and the risks from market manipulation, leverage and unsustainable positions, the
alternative is stark: we can continue to put our faith in private capital, financial and currency
markets.
References
Akerlof, G. A. (1970). The market for “lemons”: quality uncertainty and the market mechanism. Quarterly Journal of Economics, 84(3), 488–500. Retrieved from https://www.iei.liu.se/nek/730g83/artiklar/1.328833/AkerlofMarketforLemons.pdf
Ali, R. D., Haldane, A. G., & Nahai-Williamson, P. (2012). Towards a common financial language. SIFMA) Symposium: Building a Global Legal Entity Identifier Framework. New York: Bank of England. Retrieved from http://www.bankofengland.co.uk/publications/Documents/speeches/2012/speech552.pdf
Altman, D. (2013). The Dim Sum Dollar Carry Trade. Foreign Policy. Retrieved from http://www.foreignpolicy.com/articles/2013/10/29/the_dim_sum_bonds_renminbi_dollar_carry_trade
Bank for International Settlements. (2013). Triennial Central Bank Survey, (April). Retrieved from http://www.bis.org/publ/rpfx13fx.pdf
Bank of England. (2000). Bank of England Annual Report 2000. London, UK. Retrieved from http://www.bankofengland.co.uk/archive/Pages/digitalcontent/historicpubs/annualreport.aspx
Bank of England. (2012). Bank of England Annual Report 2012. Retrieved from http://www.bankofengland.co.uk/publications/Documents/annualreport/2012/2012report.pdf
Barwell, R., & Burrows, O. (2011). Growing fragilities ? Balance sheets in The Great Moderation Growing fragilities. London, UK. Retrieved from www.bankofengland.co.uk/publications/Documents/fsr/fs_paper10.pdf?
BIS. (2010). Triennial Central Bank Survey: Report on global foreign exchange market activity in 2010 (pp. 1–95). Basel, Switzerland. Retrieved from http://www.bis.org/publ/rpfxf10t.pdf
Bloomberg News. (2013). China May Limit Gold to 2% of Foreign Reserves. Retrieved from http://www.bloomberg.com/news/2013-03-13/china-may-limit-gold-to-2-of-foreign-reserves-pboc-s-yi-says.html
Bourse Consult. (2012). London : a centre for renminbi business. London, UK. Retrieved from http://www.cityoflondon.gov.uk/business/support-promotion-and-advice/promoting-the-city-internationally/china/Documents/London_A_Centre_for_RMB_business_2013.pdf
Cecchetti, S. G., Mccauley, R. N., & Mcguire, P. M. (2012). Interpreting TARGET2 balances, (393).
Chen, Z., Du, J., Li, D., & Ouyang, R. (2013). Does foreign institutional ownership increase return volatility? Evidence from China. Journal of Banking & Finance, 37, 660–669. doi:10.1016/j.jbankfin.2012.10.006
Copeland, M. (1949). Social Accounting for Moneyflows. The Accounting Review, 24(3), 254–264.
Curcuru, S. E., Thomas, C. P., & Warnock, F. E. (2008). Current Account Sustainability and Relative Reliability. New York.
Dooley, M. P., Folkerts-Landau, D., & Garber, P. M. (2009). Bretton Woods II still defines the International Monetary System. National Bureau of Economic Research. Cambridge, Mass. Retrieved from http://www.nber.org/papers/w14731.pdf
Duc, B. (2009). Flow-of-Funds analysis at the ECB. Retrieved from http://www.ecb.europa.eu/pub/pdf/scpops/ecbocp105.pdf
Ferrantino, M. J., Liu, X., & Wang, Z. (2012). Evasion behaviors of exporters and importers: Evidence from the U.S.–China trade data discrepancy. Journal of International Economics, 86(1), 141–157. doi:10.1016/j.jinteco.2011.08.006
Financial Stability Board. (2012). A Global Legal Entity Identifier for Financial Markets. Basel, Switzerland. Retrieved from http://www.financialstabilityboard.org/publications/r_120608.pdf
Gourinchas, P.-O., & Rey, H. (2007). From world banker to world venture capitalist: the US external adjustment and the exorbitant privilege. In R. H. Clarida (Ed.), G7 Current Account Imbalances: Sustainability and Adjustment (pp. 11–15). Chicago, Illinois: University of Chicago Press. Retrieved from http://www.nber.org/chapters/c0121.pdf
Hausmann, R., & Sturzenegger, F. (2006). Global imbalances or bad accounting? The missing dark matter in the wealth of nations. Centre for International Development at Harvard University. Cambridge, Mass. Retrieved from http://www.hks.harvard.edu/var/ezp_site/storage/fckeditor/file/pdfs/centers-programs/centers/cid/publications/faculty/wp/124.pdf
Hicks, J. R. (1935). A suggestion for simplifying the theory of money. Economica, 2(5), 1–19. Retrieved from http://www.jstor.org/stable/2549103
IMF Staff. (2009). The Financial Crisis and Information Gaps. Geneva, Switzerland. Retrieved from http://www.imf.org/external/np/g20/pdf/102909.pdf
Independent Evaluation Office. (2012). International Reserves: IMF Concerns and Country Perspectives. Washington D.C. Retrieved from http://www.ieo-imf.org/ieo/files/completedevaluations/IR_Main_Report.pdf
International Monetary Fund. (1993). The Framework for Monetary Statistics. Washington D.C. Retrieved from http://www.imf.org/external/pubs/ft/mfs/manual/pdf/mmfsch7.pdf
International Monetary Fund. (2008). Monetary and Financial Statistics: Compilation Guide. Washington D.C.
International Monetary Fund. (2009). Balance of Payments and International investment Position Manual, 6th Edition. Washington D.C.: International Monetary Fund. Retrieved from http://www.imf.org/external/pubs/ft/bop/2007/pdf/bpm6.pdf
James O’Toole. (2013, June 14). U.K. regulator eyes currency market manipulation. CNN Money. Atlanta, Georgia. Retrieved from http://money.cnn.com/2013/06/14/news/economy/currency-rigging/index.html
Judson, R. (2012). Crisis and Calm : Demand for U.S. Currency at Home and Abroad from the Fall of the Berlin Wall to 2011. New York. Retrieved from http://www.federalreserve.gov/pubs/ifdp/2012/1058/ifdp1058.pdf
Kinsella, S. (2011). Words to the wise: stock flow consistent modeling of financial instability, 1–14. Retrieved from http://dx.doi.org/10.2139/ssrn.1955613
Konno, S. (2010). Toward further improvement of the flow of funds accounts. Initiatives to address data gaps revealed by the financial crisis (pp. 515–525). Basel, Switzerland: Bank for International Settlements. Retrieved from http://www.bis.org/ifc/publ/ifcb34ak.pdf
Kregel, J. (2012). The Libor Scandal: The Fix Is In - The Bank of England Did It! Annandale-on-Hudson, New York.
Lawrence, R. Z. (1990). U.S. current account adjustment: an appraisal. Retrieved from http://www.brookings.edu/~/media/projects/bpea/1990 2/1990b_bpea_lawrence_hooper.pdf
Levin, S. Currency Reform for Fair Trade Act. , Pub. L. No. H.R. 639 (112th) (2011). Retrieved from http://www.govtrack.us/congress/bills/112/hr639
Murphy, B. G., Westwood, R., & Murphy, G. (2010). Data gaps in the UK financial sector : some lessons learned from the recent crisis. IFC Conference: Initiatives to address gaps revealed by the financial crisis (Vol. 34, pp. 28–44). Basel, Switzerland: Bank for International Settlements.
New York Times. (2014). Obama to Call for End to N.S.A.’s Bulk Data Collection. Retrieved from http://www.nytimes.com/2014/03/25/us/obama-to-seek-nsa-curb-on-call-data.html
Office for National Statistics. (2012a). United Kingdom National Accounts The Blue Book (Vol. 44). London, UK. Retrieved from http://www.ons.gov.uk/ons/rel/naa1-rd/united-kingdom-national-accounts/the-blue-book--2012-edition/united-kingdom-national-accounts---blue-book--2012-edition.pdf
Office for National Statistics. (2012b). Balancing the three approaches to measuring Gross Domestic Product, (July), 1–5. Retrieved from http://www.ons.gov.uk/ons/dcp171766_273489.pdf
People’s Bank of China. (2012). PBOC: Gold and Foreign Exchange Reserves Statement. Retrieved from http://www.pbc.gov.cn/publish/html/kuangjia.htm?id=2012s09.htm
Reserve Bank of India. (2010). Balance of payments manual for India. Retrieved from http://rbidocs.rbi.org.in/rdocs/Publications/PDFs/IBPM221110P2.pdf
Sinn, H.-W., & Wollmershäuser, T. (2011). Target Loans, Current Account Balances and Capital Flows : The ECB’s Rescue Facility. Munich, Germany. Retrieved from http://www.cesifo-group.de/DocDL/cesifo1_wp3500.pdf
Sub-committee on National Income Statistics. (1947). Measurement of National Income and the construction of social accounts. Geneva, Switzerland: United Nations. Retrieved from http://unstats.un.org/unsd/nationalaccount/docs/1947NAreport.pdf
Swiss National Bank. (2013a). Foreign currency investments of the SNB by currency. Zuich, Switzerland.
Swiss National Bank. (2013b). Banknote circulation. Retrieved August 16, 2013, from http://www.snb.ch/en/iabout/cash/id/cash_circulation
United Nations. (1953). A System of National Accounts. New York: United. Retrieved from http://unstats.un.org/unsd/nationalaccount/docs/1953SNA.pdf
US Department of the Treasury. (2013). Holdings of U.S. Long-term Securities by Foreign Residents. Retrieved from http://www.treasury.gov/resource-center/data-chart-center/tic/Documents/slt1f_globl.csv
Various. (2008). System of National Accounts. New York. Retrieved from http://unstats.un.org/unsd/nationalaccount/docs/SNA2008.pdf
Vitali, S., Glattfelder, J. B., & Battiston, S. (2011). The network of global corporate control. PloS ONE, 6(10), e25995. doi:10.1371/journal.pone.0025995
Wang, P. (2005). The Economics of Foreign Exchange and Global Finance. Berlin, Heidelberg, New York: Springer.
Xu, Y., Lin, G., & Sun, H. (2010). Accounting for the China-US Trade Imbalance: An Ownership-Based Approach. Review of International Economics, 18(3), 540–551. doi:10.1111/j.1467-9396.2010.00882.x