New feature: Realistic Loan Demand

[David]

In the latest post-release beta version 6.5.00, a new option called “Realistic Loan Demand” is available on the “Bank” page of the New Game Setting menu.

Download link: http://www.capitalism2.com/forum/viewto ... =10&t=7665

When Realistic Loan Demand is enabled:
a) The demand for bank loans is correlated to the total GDP of all cities.

b) The game shows a new setting on the “Deposits from HQ” page of the Bank HQ interface called “Threshold for Stopping Accepting Corporate Deposits” If this is enabled, the bank stops accepting new deposits from corporate clients when the bank's cash as a percentage of its total assets exceeds the specified threshold.

[buells]

As it is currently set up, this feature simply doesn't work well. At the beginning of the game a bunch of banks are set up and satisfy all of the loan demand. In real life there is always demand for loans, like any other product. Demand for loans follows a curve like anything else. In a depression there may realistically be no loan demand, but in a boom, a new bank always has the opportunity to COMPETE to lend.

The demand system should not be based on the "stock" of loans, rather there is a constant demand for new loans to be originated every day. If a bank has $2bn in deposits, they can ALWAYS issue loans, but they may need to offer loans with lower margin (implying a lower net interest margin for them) to capture share from incumbents. Even if they have to issue loans with lower rates than their cost of deposits, they aren't going to sit around paying interest on deposits on which they earn absolutely nothing.

[This part is explanatory, skip past next brackets to get to my concrete suggestions.]

To say that loan demand is limited to the GDP of a given place is also incorrect. It is CORRELATED to GDP, to an extent, to be sure, but loans create most of the money used in the economy. The money is used for transactions (captured in GDP) but also for creating financial assets. For example, if a company takes out a loan, that creates money but it isn't part of GDP. If they use the loan to buy land, that isn't included in GDP either. The seller of the land now has money that they may hold in bank deposits for a period of time before disbursing it to buy other assets (not necessarily part of GDP), to invest in a new factory (part of GDP), or to buy consumer goods (part of GDP).

My intention in suggesting previously that loan demand should factor into the game was driven by the fact that in reality banks do not receive deposits and then go lend them out. They arrange to make loans and then fund them. Initially the whole balance of the loan may be deposited into the originating bank (not always, obviously), in which case it is initially self-funding. Banks generally keep a cushion of funding such that issuing loans in the normal course of business does not require them to go out and seek more deposits. On a daily basis, their treasury functions ensure enough deposits or wholesale funding is available to make loans and keep some cushion. They seek more deposits because these have a cheaper funding cost than wholesale funding for a given maturity. They adjust the interest rates they are offering for new deposits as needed to maintain their targeted liquidity. Making loans is a harder part of the business because they have to underwrite them to assess credit risk (among other risks like fraud, money laundering, etc.) In theory they could issue an unlimited amount of loans (people will always want loans if they cost nothing), but they don't do this because they need to earn a targeted return on their equity capital. The amount of loans supplied in a given day, week, month, or year will depend on the intersection of the supply and demand curve (the equilibrium of loan quantity and interest rate that describes a given market). The demand for loans obviously declines with the interest rate. In a competitive credit market, banks will be more willing to supply loans as interest rates rise. The supply curve is defined by the costs of the bank to make a loan (principally underwriting expenses, expected credit losses, and cost of funds, including cost of equity capital required to maintain capital requirements).

[Suggestions continue below]

The reality is, the actual financial markets are somewhat more complex than this and have a feature of reflexivity, which is important. However, for the purposes of a simulation, the foregoing description is pretty accurate. My thinking is that loans should be automatically issued in the game at a rate determined by a formula based on (A) bank branch service level, (B) city bank brand rating, (C) status of the economy, and (D) loan interest rate (defined as margin over base rate). The Total amount of loans you can have outstanding at any one time should be limited by bank equity capital ratio as it is now. It should NOT be limited by deposits. New loans should be automatically funded with wholesale financing. The amount of deposits you have should automatically replace wholesale financing as they accumulate, which would benefit the player by being a cheaper funding source. Deposits in excess of your stock of commercial loans should automatically be lent out as at a low interest rate (figure something at about the average rate paid on short term commercial deposits by the all the banks at a given time). No other changes are needed to the deposit mechanism. This also opens the door to other gameplay features for the lending process.

Having different kinds of lending franchises within your bank (perhaps auto, credit card, mortgage, and general commercial) can create different market segments with different economic correlations in which different banks could be stronger than others. I would think you'd have a separate brand for each and perhaps a slider for investment into underwriting capability for each, which would help drive market share and subsequent credit losses. Auto loan growth could correlate to auto sales, credit cards to retail sales, and mortgages/commercial to GDP growth. Auto and credit card defaults would rise more sharply when the economy tanks like (they could simply flow into one of the lower credit rating tiers upon origination). This is a nice to have rather than a need to have, but it would be a way to significantly expand the content provided by the DLC.

Anyway, I am not hugely concerned if my suggestions in the last two paragraphs don't get implemented, but I just want to point out that the current "realistic loan demand" feature does not work.

[David]

FYI, the loan demand was implemented based on the user suggestions in this discussion thread: viewtopic.php?f=52&t=7510

Here is an excerpt of the discussions:

You can do something simple like private credit as % of GDP (which globally/historically is in the 100-200% range). If you want to get cute about it, you can add some modifiers that basically make it grow/shrink with recession/expansion and generally show a positive correlation with relative wealth - richer countries do tend to have more debt (markets are more stable / predictable so creditors can lend more). So in the US, for instance, private credit is ~200% of GDP and it grows/drops with expansions/recessions.

[buells]

That's all well and good, but the way it is currently implemented you can't take market share from competitors. Loan demand for new loans should be a fixed amount at any one time, but the stock of loans should not be effectively capped.